Russell "On Denoting" [Adam Plesser]

I agree with Russell's distinction between primary and secondary occurrences of denoting phrases. I do not believe there is a satisfactory way of making sense of the statement "the king of France is bald" that does not utilize a distinction like Russell's. This is because the statement is clearly not true, but it can only be considered false if understood in a certain way and therefore there must be some way of separating the phrase "it is false that the king of France is bald" into at least two different ways in which it can be understood.

I disagree with Russell's insistence on structuring denoting phrases in terms of truth values. If, for example, we're crossing the street with a friend and we say "car!" what we are saying is not "it is true that there is a car" or some such thing, we're saying "stop walking" or "step back onto the sidewalk." Perhaps even this could be transformed into "it is true that if you continue walking, you will be run over" but this is not actually what we're saying. The former phrases are all that we are saying while this latter interpretation lumps in our motivations, which we are not actually expressing, they are simply clear to us as objective observers. This means that either Russell is wrong to frame his understanding of denoting phrases in terms of truth values or he is wrong to define denoting phrases such that this case counts as an instance of a denoting phrase.

I don't understand the difference between the denoting phrase "the difference between A and B" in cases where A and B are identical and the denoting phrase "the king of France" or any other denoting phrase that does not denote anything. Is there a difference? Why did he discuss these cases separately?